Complement C5 inhibitors as add-on rescue treatment in severe AQP4-antibody neuromyelitis optica spectrum disorder attacks: a case series and literature review

Acute attacks of aquaporin-4 antibody neuromyelitis optica spectrum disorder (AQP4-Ab NMOSD) are typically managed with intravenous methylprednisolone, with therapeutic plasma exchange (PLEX) reserved for severe episodes. Despite these interventions, a subset of patients fails to respond and may sustain significant neurological disability. Complement C5 inhibitors, currently approved for long-term relapse prevention, have recently gained attention as a potential rescue therapy by rapidly inhibiting the terminal complement cascade, thus preventing downstream astrocytic injury. We report three patients with AQP4-Ab severe optic neuritis refractory to IV methylprednisolone and PLEX who demonstrated significant clinical improvement following administration of C5 inhibitors during the acute attack phase. Case 1 (right optic neuritis) and Case 3 (bilateral optic neuritis) received eculizumab, achieving near-complete and complete recovery of visual acuity, respectively. Both patients subsequently transitioned to rituximab within 8 weeks after their fourth weekly eculizumab infusion, while on a moderate dose of prednisolone. Case 2 (left optic neuritis) received ravulizumab, achieving partial visual recovery, and was continued on maintenance ravulizumab. Meningococcal prophylaxis with penicillin V was administered for all three patients. A literature review identified 10 publications utilising C5 inhibitors as acute attack therapies, totalling 31 cases, including our present series. Of these, 30 patients (97%) experienced neurological improvement within 4 weeks, with 28 (90%) showing improvement within the first week. Complete or near-complete recovery was observed in 42% of patients, while 55% achieved partial recovery, and one patient (3%) had no recovery. Patients achieving complete or near-complete recovery received C5 inhibitors significantly earlier than those with partial or no recovery (median 15.0 vs 24.0 days from symptom onset, p = 0.010); an optimal analytical cutoff of 23 days was identified. Infections were uncommon with no severe infections reported. In conclusion, emerging evidence supports the use of C5 inhibitors for acute NMOSD attacks, but larger studies are needed to establish robust evidence and inform clinical practice.